Abstract
The main research question of this article was whether different reduced word forms are stored in the mental lexicon of the speaker. This was answered by means of a production experiment where participants coming from Roosendaal and Ede had to read phrases that were presented to them. These phrases contained two nouns, the first ending in the cluster /st/ and the second beginning with a consonant.
Reduced word forms were measured and results show that the effect of dialect were strongly significant with regard to the pronunciation of /s/ (F(1, 1022) = 78.399; p < .001) and /t/ (F(1, 1022) = 736.293; p < .001). Specifically, people speaking the dialect of Ede pronounced /t/ in such a way that it added up to a total of 5% of the total word while this was 16% for the people of Roosendaal.
The final experiment contained a lexical decision task and it showed that people from Ede were a lot less accurate when compared to speakers of Roosendaal (41.66% less accurate, to be exact). This means that people from Ede often perceived reduced words as words. We propose a hybrid-model of word recognition.
Reduced word forms were measured and results show that the effect of dialect were strongly significant with regard to the pronunciation of /s/ (F(1, 1022) = 78.399; p < .001) and /t/ (F(1, 1022) = 736.293; p < .001). Specifically, people speaking the dialect of Ede pronounced /t/ in such a way that it added up to a total of 5% of the total word while this was 16% for the people of Roosendaal.
The final experiment contained a lexical decision task and it showed that people from Ede were a lot less accurate when compared to speakers of Roosendaal (41.66% less accurate, to be exact). This means that people from Ede often perceived reduced words as words. We propose a hybrid-model of word recognition.
Publication type
Presentation
Year of publication
2014
Conference location
Utrecht
Conference name
Dag van de Fonetiek 2014
Publisher
Nederlandse Vereniging voor Fonetische Wetenschappen